Research On Optimization Of Cascaded Model Predictive Control System For Linear Motor

Currently,China is in a period of rapid development in industrial technology,and there is an increasing demand for linear drive equipment in modern industry.Permanent magnet synchronous linear motors are widely used in the industry due to their high precision and efficiency.The modern industry has increasingly high requirements for the dynamic and steady-state performance of direct drive equipment,and model predictive control strategies are increasingly applied in the field of motor speed regulation.However,model predictive control requires high accuracy of the parameters,and when the parameters change,the performance of the control system is affected.Therefore,improving the robustness of the model predictive control system is crucial.Firstly,a model predictive cascade speed regulation system was designed,and then,methods were proposed to enhance the robustness of the current loop and speed loop respectively,in order to address the robustness issues of the model predictive speed regulation system.An incremental model predictive control strategy based on perturbation observation and inductance correction is proposed in the current loop.Firstly,the parameter sensitivity analysis of existing model predictive current control is carried out,and the influence of system parameters on the predicted value is quantified.The prediction model is changed to incremental prediction model to eliminate the influence of flux in the prediction model.Then a disturbance observer is established to compensate the disturbance into the incremental prediction model,and the actual inductance value is calculated through the disturbance to adjust the inductance parameters of the prediction model.Simulation results show that this method can improve the performance of the system from two aspects:making up for the model error and improving the model precision.In the speed loop,Luenberger load observation and internal model control load observation are proposed to suppress the load disturbance.In the traditional model predictive control,the disturbance term is ignored and only the feedback correction link is used to suppress the disturbance.The inhibition effect is limited.To solve this problem,different load observers based on velocity loop are constructed to observe the load disturbance in the system,and the disturbance is suppressed according to the reference current compensation generated by the observed load.The simulation results show that adding load observation feedforward compensation can enhance the anti-disturbance capability of the system.Finally,the software and hardware experimental platform of the control system is built.Through the motor speed regulation,parameter mismatch and load surge experiments,it is verified that the predictive control system designed in this paper not only has good dynamic performance,but also has high parameter robustness and anti-disturbance ability.